Deep learning model based on primary tumor to predict lymph node status in clinical stage IA lung adenocarcinoma:a multicenter study

Objective:To develop a deep learning model to predict lymph node(LN)status in clinical stage IA lung adeno-carcinoma patients.Methods:This diagnostic study included 1,009 patients with pathologically confirmed clinical stage T1N0M0 lung adenocarcinoma from two independent datasets(699 from Cancer Hospital of Chinese Academy of Medical Sciences and 310 from PLA General Hospital)between January 2005 and December 2019.The Cancer Hospital dataset was randomly split into a training cohort(559 patients)and a validation cohort(140 patients)to train and tune a deep learning model based on a deep residual network(ResNet).The PLA Hospital dataset was used as a testing cohort to evaluate the generalization ability of the model.Thoracic radiologists manually segmented tumors and interpreted high-resolution computed tomography(HRCT)features for the model.The predictive performance was assessed by area under the curves(AUCs),accuracy,precision,recall,and F1 score.Subgroup analysis was performed to evaluate the potential bias of the study population.Results:A total of 1,009 patients were included in this study;409(40.5%)were male and 600(59.5%)were female.The median age was 57.0 years(inter-quartile range,IQR:50.0-64.0).The deep learning model achieved AUCs of 0.906(95%CI:0.873-0.938)and 0.893(95%CI:0.857-0.930)for predicting pN0 disease in the testing cohort and a non-pure ground glass nodule(non-pGGN)testing cohort,respectively.No significant difference was detected between the testing cohort and the non-pGGN testing cohort(P=0.622).The precisions of this model for predicting pN0 disease were 0.979(95%CI:0.963-0.995)and 0.983(95%CI:0.967-0.998)in the testing cohort and the non-pGGN testing cohort,respectively.The deep learning model achieved AUCs of 0.848(95%CI:0.798-0.898)and 0.831(95%CI:0.776-0.887)for predicting pN2 disease in the testing cohort and the non-pGGN testing cohort,respectively.No significant difference was detected between the testing cohort and the non-pGGN testing cohort(P=0.657).The recalls of this model for predicting pN2 disease were 0.903(95%CI:0.870-0.936)and 0.931(95%CI:0.901-0.961)in the testing cohort and the non-pGGN testing cohort,respectively.Conclusions:The superior performance of the deep learning model will help to target the extension of lymph node dissection and reduce the ineffective lymph node dissection in early-stage lung adenocarcinoma patients.

Reappraise role of lymph node status in patterns of recurrence following curative resection of gastric adenocarcinoma

Objective: To examine the association between lymph node status and recurrence patterns in completely resected gastric adenocarcinoma.Methods: We retrospectively assessed 1,694 patients who underwent curative gastrectomy from January 2010 to August 2014. Patients stratified according to lymph node status and recurrence patterns among different subgroups were compared.Results: Of all, 517(30.5%) patients developed recurrent disease, and complete data of recurrence could be obtained in 493(95.4%) patients. For p^(N0) patients, the patterns of recurrence were different according to p T stage: locoregional recurrence was most common in patients with p T1-2 disease(57.1%), distant recurrence was most common in patients with p T3 disease(57.1%), and peritoneal recurrence was most common in patients with p T4 a disease(66.7%). For p^(N+) patients, distant metastasis was most common pattern irrespective of p T stage. The site-specific trend of recurrence showed that locoregional recurrence increased within 5 years in patients with p^(N0)-2 disease but plateaued 3 years after surgery in patients with p N3 disease. Time to recurrence was significantly longer for the p^(N0) patients compared with the p^(N+) patients(median: 25 vs. 16 months, P=0.001).Moreover, post-recurrence survival was significantly better for the p^(N0) patients than for the p^(N+) patients(median:12 vs. 6 months, P<0.001), especially in patients with non-peritoneal recurrence, late recurrence, single recurrence,and receipt of potential curative treatment.Conclusions: Among clinicopathologic factors, lymph node status is the most important factor associated with recurrence patterns after curative gastrectomy. Lymph node status may be used as an adjunct in clinical decisionmaking about postoperative therapeutic and follow-up strategies.

Primary Treatment for Clinically Early Cervical Cancer with Lymph Node Metastasis:Radical Surgery or Radiation?

Objective To compare survival outcomes between primary radical surgery and primary radiation in early cervical cancer.Methods Patient information was extracted from the Surveillance,Epidemiology,and Results database.Patients diagnosed with early cervical cancer of stage T1a,T1b,and T2a(American Joint Committee on Cancer,7th edition)from 1998 to 2015 were included in this study after propensity score matching.Overall survival(OS)was analyzed using the Kaplan-Meier method.Results Among the 4964 patients included in the study,1080 patients were identified as having positive lymph nodes(N1),and 3884 patients were identified as having negative lymph nodes(N0).Patients with primary surgery had significantly longer 5-year OS than those with primary radiotherapy in both the N1 group(P<0.001)and N0 group(P<0.001).In the subgroup analysis,similar results were found in patients with positive lymph nodes of stage T1a(100.0%vs.61.1%),T1b(84.1%vs.64.3%),and T2a(74.4%vs.63.8%).In patients with T1b1 and T2a1,primary surgery resulted in longer OS than primary radiation,but not in patients with T1b2 and T2a2.In multivariate analysis,the primary treatment was identified as an independent prognostic factor in both N1 and N0 patients(HR_(N1)=2.522,95%CI=1.919–3.054,PN1<0.001;HR_(N0)=1.895,95%CI=1.689–2.126,PN0<0.001).Conclusion In early cervical cancer stage T1a,T1b1,and T2a1,primary surgery may result in longer OS than primary radiation for patients with and without lymph node metastasis.

Effect of the number of positive lymph nodes and lymph node ratio on prognosis of patients after resection of pancreatic adenocarcinoma

BACKGROUND： The prognostic factors related to lymph node involvement [lymph node status, the number of positive lymph nodes, lymph node ratio （LNR）] and the number of nodes evaluated in patients with pancreatic adenocarcinoma after pancreatectomy are poorly defined. METHODS： A total of 167 patients who had undergone resection of pancreatic adenocarcinoma from February 2010 to August 2011 were included in this study. Histological examination was performed to evaluate the tumor differentiation and lymph node involvement. Univariate and multivariate analyses were made to determine the relationship between the variables related to nodal involvement and the number of nodes and survival. RESULTS： The median number of total nodes examined was 10 （range 0-44） for the entire cohort. The median number of total nodes examined in node-negative （pN0） patients was similar to that in node-positive （pN1） patients. Patients with pN1 diseases had significantly worse survival than those with pN0 ones （P=0.000）. Patients with three or more positive nodes had a poorer prognosis compared with those with the negative nodes （P=0.000）. The prognosis of the patients with negative nodes was similar to that of those with one to two positive nodes （P=0.114）. The median survival of patients with an LNR ≥0.4 was shorter than that of patients with an LNR 〈0.4 in the pN1 cohort （P=0.014）. No significance was found between the number of total nodes examined and the prognosis, regardless of the cutoff of 10 or 12 and in the entire cohort or the pN0 and pN1 groups. Based on the multivariate analysis of the entire cohort and the pN1 group, the nodal status, the number of positive nodes and the LNR were all associated with survival. CONCLUSIONS： In addition to the nodal status, the number of positive nodes and the LNR can serve as comprehensive factors for the evaluation of nodal involvement. This approach may be more effective for predicting the survival of patients with pancreatic adenocarcinoma after pancreatectomy.

Pure Mucinous Carcinoma of the Breast:a Clinicopathologic Analysis with 56 Patients

Objective To assess recent trends and prognostic features in the treatment of pure mucinous carcinoma(PMC) of the breast.Methods Fifty-six patients diagnosed with PMC of the breast in our hospital from December 1982 to June 2008 were included.We evaluated the general information and tumor characteristics of the patients,examined the relationship between these factors and prognosis.Fisher's exact test was applied to analyze tumor characteristics.Results The mean age of the patients was 53.7 years.The majority of the patients presented with early stage disease.Tumor size was found not a significant prognostic factor in this study.Mean follow-up period was 39 months and no breast cancer-related deaths were identified in the patient cohort.Conclusions PMC of the breast has a favorable prognosis.Tumor size does not appear to significantly impact survival.